Corticosteroid-related toxicity in patients with chronic idiopathic urticariachronic spontaneous urticaria.

BACKGROUND Treatments for patients with chronic idiopathic urticaria (CIU)chronic spontaneous urticaria (CSU) who were unresponsive to antihistamines include oral corticosteroids (OCS). Risks of OCS-related side effects in these patients have not been described quantitatively. OBJECTIVE To investigate the relationship between OCS use and the risk of developing side effects possibly attributable to OCS and associated health care costs in privately insured patients with CIU/CSU. METHODS This retrospective cohort study analyzed a commercial claims data base from January 1, 2008, to December 31, 2012. Patients with CIU/CSU were identified by International Classification of Diseases, Ninth Revision, Clinical Modification codes via a validated algorithm. Possible OCS-related side effects included the following: diabetes mellitus, hypertension, lipid disorders, cataracts, depression or mania, osteoporosis or fractures, and infectious diseases. A time-dependent Cox regression (adjusted for age, sex, Charlson Comorbidity Index, and immunomodulator use) was used to separately model cumulative oral prednisone-equivalent exposure and the risk of side effects. Incremental total adjusted health care costs were compared in patients with versus patients without possible OCS-related side effects. RESULTS Among 12,647 patients with CIU/CSU, 55.4% used OCS. An additional 1 g of prednisone-equivalent exposure was associated with a 7% increase in the likelihood of developing a possible side effect (hazard ratio, 1.07 [95% confidence interval, 1.051.08]). From the period before to the period after OCS initiation, the total mean adjusted annual health care costs increased by 1833 in users of OCS with new possible side effects and decreased by 2183 in patients without new possible side effects (p 0.001). CONCLUSION Patients with CIU/CSU who were treated with OCS had an increased risk of possible OCS-related side effects and higher total health care costs than their counterparts not treated with OCS.